Morphometric analysis of suprabasal cell layer in oral epithelial dysplasia: A computer-assisted microscopic study

Abstract

A BSTRACT Background: Oral leukoplakia is the most common precancerous lesion. Various grading systems based on histological findings have been proposed for dysplasia. Recently, computer-assisted morphometric analysis has been established to reduce the interobserver and intraobserver variability during the histological grading of epithelial dysplasia. This study was undertaken to establish the morphometric changes in the suprabasal cell layer of different grades of oral epithelial dysplasia. Materials and Methods: Forty paraffin-embedded tissue sections (10 normal mucosa, 10 mild dysplasia, 10 moderate dysplasia, and 10 severe dysplasia cases) were stained using hematoxylin and eosin stain, and analyzed for cellular and nuclear morphometry using binocular microscope. Results: Our results showed that values of nuclear area, nuclear perimeter, nuclear volume density, and nuclear/cytoplasmic (N/C) ratio were increased gradually in dysplasia compared to control groups and the values were statistically significant ( P = 0.001). Nuclear diameter and cellular area were increased in dysplasia when compared to control group ( P = 0.001). Mild and moderate dysplasia showed decreased value of nuclear form factor compared to control group, whereas severe dysplasia showed highest value. A fair correlation was found when comparing histological grading and grouping based on nuclear area, nuclear perimeter, N/C ratio, and nuclear volume density. Conclusion: Nuclear features reflect cell behavior, and its morphometric analysis can be considered as a reliable tool for differentiating various grades of epithelial dysplasia.

The use of deep learning state‐of‐the‐art architectures for oral epithelial dysplasia grading: A comparative appraisal

“…The authors acknowledge that support systems for OED grading should overcome this percentage but particularities in the data curation and the investigated disorders should be considered. A statistically significant increase in nuclear volume density, nuclear‐cell ratio, nuclear area, and perimeter are found in dysplastic cells when compared within different dysplasia grades, with higher values in increasing grades 28 . This premise corroborates that shape and size of nuclei can help in the differentiation among dysplasia grading.…”

“…A statistically significant increase in nuclear volume density, nuclear-cell ratio, nuclear area, and perimeter are found in dysplastic cells when compared within different dysplasia grades, with higher values in increasing grades. 28 LR. However, annotations enrolling all the epithelium extension associated with the fragmentation process may insert bias by providing a great intersection of features that characterize both OED classes when using the BS grading system.…”

The Role of Cell Detection Tool of QuPath-0.3.2 Software in Differentiating Benign Cervical Epithelium from Malignant Epithelial Clusters

“…Many researchers have tried to use image analysis to improve tissue diagnosis of cancer [1][2][3][4]. There has been a lot of research that showed the possible role of computer-aided cell morphometry in diagnosis of lesions and predicting clinical outcomes [5][6][7][8]. Most of these works studied nuclear morphometry in terms of nuclear size and area.…”